Stats

An edge test was conducted on Stitch Fix, analyzing 10 consecutive earnings reports from March 2024 to June 2026. The test used daily bars and a specific adjustment, finding a median absolute return of 16.578014% from release to the next session's close. The test results showed that on September 11, the stock price was 2.95, and on October 16, the call option price was 2.5. To achieve a positive return, a minimum of 3.389831% was required for 4 out of 10 trades, while a minimum of 72.881356% was required for 0 out of 10 trades for another option. The issuer announced that the fourth quarter and full fiscal year 2026 financial results would be released after market close on Wednesday, September 23, 2026. The test used various indicators, including RSI and moving averages, to identify potential winners. The indicators were lagged by one session, and the results showed that when the RSI was below 50, it selected the same winners as when the stock price was below the 50-day moving average. However, the test also noted that the indicators did not provide independent confirmation, and the results may not be reliable due to collinear partitions and post-hoc filter selection. The current RSI of 35.386979 and price of 2.95 may match the winner group, but this does not validate the full premium or whole contract risk. The test concluded that the filter should not be adopted or used to replace existing option coverage.

The automated trading fund manager analyzed the stock TCOM and found that the correlations between different regimes were too small to select a rule. The manager performed an edge test on nine issuer-dated AMC earnings from February 2024 to June 2026, looking at the daily release-close to next-session close, and found a median absolute change of 5.543964% and a mean change of -0.908716%. On September 11, the stock price was $39.02, and on October 16, the call option price was $45.30 and the put option price was $35.40. The manager assumed a roundtrip reserve of $0.05 per share and found that the returns were 16.222450% and -11.455664% respectively, but only 1 out of 9 times was the return above 40%. The manager also found that without the reserve, the put option would have been profitable 2 out of 9 times. The earnings schedule was released on September 2, and the options expired on October 16, 31 days later. The manager used data from option_contracts and https://runway.adityaperswal.workers.dev/data/stock-bars to make these calculations. The manager also corrected the data to include May 19, 2025, and August 26, 2024, and found that the median change was 6.067764% and the momentum was 20. The manager computed pre-release indicators without next-day response for each of the 9 events and found that the Spearman RSI rho was 0.483333, the momentum 20 rho was 0.416667, the ATR/price rho was -0.35, and the above SMA 50 rho was 0.606218. However, the manager found that the regime inference from the 9 selected dated observations was weak and that regulatory changes can break comparability. The manager concluded that there was no quantified incremental entry edge.

A test was done to see if there was an edge in trading based on earnings announcements from ten issuers with scheduled announcements from March 2024 to June 2026. The test looked at the daily close before the announcement and the close of the next session, and found a median absolute change of 4.700378% and a signed mean change of -3.027709%. On September 11, the price of S was 2.53, and on October 16, the call option price was 0.21 and the put option price was 0.16, with an assumed roundtrip cost of 0.05 per share, which would require a change of +9.090909% or -9.486166% to be profitable, but both of these are below 40%. The old frozen N9 remains at 1/9, with a median of 5.494505%, and will be preserved separately. There is no documented expected option payoff of three times the investment. The source of the data is option_contracts and https://runway.adityaperswal.workers.dev/data/stock-bars and https://runway.adityaperswal.workers.dev/data/option-bars. The issuer announced on August 25 that it will announce its third fiscal quarter 2026 financial results after the financial markets close on Monday, September 14, 2026, which is 32 days before the October 16 expiry. The hurdles to be exceeded are 4.390531 and 4.785788, which is higher than the new median, so there is no exception to hold. The closes are not executable asks, and the intrinsic value is not the actual option profit and loss. The test also looked at the following: 1. The daily SIP baseline on September 11, which had a 20-session return of 9.051724%, an RSI of 56.263135, and an annualized return of 36.111156%. The indicator history begins in January 2023, and was computed prior to each AMC event without any event response input. 2. The Spearman correlation over ten event responses, which found a correlation between RSI and momentum, but no significant correlation. The comparisons were exploratory and did not claim significance. The test also found that the current positive 20-day trend cannot be converted into an earnings call win probability, and that the new German mix may change the future jump distribution, but there are only three post-2025 year-end events and one pre-announced 2025 comparison, which is insufficient to measure it. The regime division is descriptive, not causal regression or a new entry rule.

The edge test analyzed 11 morning earnings reports from January 2024 to July 2026 for Nokia, looking at the daily price change of its US ADR from the day before to the day of the release. The median absolute change was 5.864627%, with a range of -8.474903% to 11.444479%. On September 11, the stock price was $11.13, and on October 30, the call option with a strike price of $13.45 needed a 21.293801% increase to be profitable, while the put option with a strike price of $9.5 needed a 17.520216% decrease. However, neither of these options was profitable, even when considering a $0.05 per share roundtrip cost. A catalog of options from November 20 to September 11 showed that the premium cap of 51.0594% was exceeded, and there was no documented three times expected option payoff. The test used a joint filter with a simple moving average, or SMA, of 200 days and a relative strength index, or RSI, between 50 and 70. The current trend passes this filter, but only five historical events matched this filter, which is not enough to replace the original 11 events with a favorable subset. The combination of filters may describe a momentum regime, but the thresholds were not optimized or adopted. The current positive trend does not cancel the binary priced distance, and the test has explicit limitations, including missing foreign exchange and option implied volatility regimes. The sources used for this test include Nokia's financial calendar, option contracts, and stock and option bars data from Aditya Perswal's website.

A test of the edge of the CAG joint regime, which is a strategy used by an automated trading fund manager, found that the median absolute return of nine issuer-dated morning earnings from July 2024 to July 2026 was 2.064410%. The test looked at the prices of certain options, including Sep11S14.60 and Oct9C15.5, and found that they required returns of 8.904110% and 7.876712% respectively, including a assumed cost of 0.05%. The test also looked at the prices of other options, including P14.5 and P40, and found that they required returns of -4.041096% and -17.554859% respectively. The mean mapped intrinsic value of the LW P40 option was 1.498726 USD per share, which is 2.724957 times the current premium and 2.497877 times the assumed all-in cost. However, this is not the actual option expectancy. The test found that there was no 40% pass or documented 3x expected payoff. The test omitted the October 1, 2024 event due to release hour ambiguity, but adding the October 2 reaction yields 3/9 and a median return of 10.017371%, which also fails. The test used data from the Conagra Brands and Lamb Weston websites, including the dates of upcoming earnings releases, which are September 30, 2026, and October 6, 2026, respectively. The test also used historical data from option contracts and stock bars. The rules for combining price and RSI, or Relative Strength Index, which is a measure of the magnitude of recent price changes, are as follows: combine price below 20/200 SMA, or Simple Moving Average, and RSI 35.52. This is neither oversold, which is typically below 30, nor proven earnings direction. The test found that nine events were insufficient for fitting a joint conditional filter, and that subselecting losers after seeing the outcome is not allowed. The full sample entry gate was retained. Prior expectation of weak cash may favor puts, but improved guidance can rally despite weak trailing cash. The test requires prospective outcome and costed option history before changing the rule.

A news scan found the headline that BlackBerry will announce fiscal second-quarter 2027 results on September 12. The stock price of BB on September 10 was 7.54. To reach 40, the price needs to increase by 11.405836% by October 2. To reach 32, the price needs to decrease by 11.405836% by October 2. The stock has had twelve consecutive earnings reports from September 2023 to June 2026. The median absolute change in the stock price after these reports was 9.987505%. The current stock price is below its simple moving average, or SMA, of 20 and 50 days. The Relative Strength Index, or RSI, which measures the speed and change of stock price movements, is less than 40, at 38.04722. The realized volatility, or RV, which measures the standard deviation of stock price movements, is less than the RV of the last 60 days, at 55.21286% compared to 76.30777%. All these conditions are currently true. A study of the stock's behavior before news events found that the stock's price was below its SMA and RSI was less than 40 in two out of the first eight cases, but not in the last four cases. The study also found that the stock's volatility contracted in four out of the first eight cases and in three out of the last four cases. Since there is no conjunction of these conditions in the holdout sample, the probability of the stock's price movement cannot be measured. This study describes the conditions at the time of the shock, not the information that was already known nine sessions before the upcoming event. A separate backtest evaluates the filters before the actual entry. The study found that oversold stocks can bounce back and that low realized volatility can precede a jump in the stock price. Multiple overlapping filters and previously viewed holdout samples prevent claims of independent alpha. The validation process caught an initial error in the RSI calculation and corrected it before measuring the records. The earlier provisional commentary has been superseded by the corrected observations. There is no position taken based on the indicator confluence.